Datasets and R Markdown files for the article "Survey on critical results management in Brazilian clinical laboratories: Profiling practices through multivariate analysis, prioritization, and a 'New Statistics' approach" submitted to Clinica Chimica Acta

This repository contains supplementary materials related to the study "𝐒𝐮𝐫𝐯𝐞𝐲 𝐨𝐧 𝐜𝐫𝐢𝐭𝐢𝐜𝐚𝐥 𝐫𝐞𝐬𝐮𝐥𝐭𝐬 𝐦𝐚𝐧𝐚𝐠𝐞𝐦𝐞𝐧𝐭 𝐢𝐧 𝐁𝐫𝐚𝐳𝐢𝐥𝐢𝐚𝐧 𝐜𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐥𝐚𝐛𝐨𝐫𝐚𝐭𝐨𝐫𝐢𝐞𝐬: 𝐏𝐫𝐨𝐟𝐢𝐥𝐢𝐧𝐠 𝐩𝐫𝐚𝐜𝐭𝐢𝐜𝐞𝐬 𝐭𝐡𝐫𝐨𝐮𝐠𝐡 𝐦𝐮𝐥𝐭𝐢𝐯𝐚𝐫𝐢𝐚𝐭𝐞 𝐚𝐧𝐚𝐥𝐲𝐬𝐢𝐬, 𝐩𝐫𝐢𝐨𝐫𝐢𝐭𝐢𝐳𝐚𝐭𝐢𝐨𝐧, 𝐚𝐧𝐝 𝐚 '𝐍𝐞𝐰 𝐒𝐭𝐚𝐭𝐢𝐬𝐭𝐢𝐜𝐬' 𝐚𝐩𝐩𝐫𝐨𝐚𝐜𝐡". The dataset, figures, exported results, and analysis scripts are included to ensure full transparency and reproducibility of the research findings.

𝐅𝐨𝐥𝐝𝐞𝐫 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞

1_𝐃𝐚𝐭𝐚𝐬𝐞𝐭/
This folder contains the dataset used in the study, formatted for direct use in the Feature Priorizer R Markdown script.

2_𝐅𝐢𝐠𝐮𝐫𝐞𝐬/
All figures generated by the Feature Priorizer are stored here in 600 DPI resolution, ensuring high-quality graphics for publication and analysis.

3_𝐄𝐱𝐩𝐨𝐫𝐭𝐞𝐝/
This folder contains the exported results, including statistical outputs, tables, and processed datasets derived from the analyses.

4_𝐒𝐮𝐩𝐩𝐥𝐞𝐦𝐞𝐧𝐭𝐚𝐫𝐲_𝐅𝐢𝐥𝐞𝐬/
This folder contains auxiliary files used in generating the Feature Priorizer HTML report, ensuring an enhanced visual presentation and incorporating dynamic statistical quotes.
– 𝐬𝐭𝐲𝐥𝐞𝐬.𝐜𝐬𝐬: Defines the formatting of the HTML report, ensuring a consistent visual presentation.
logo.html, logo.png, logo.txt – Files related to the project's visual identity.
– 𝐒𝐜𝐢𝐞𝐧𝐜𝐞_𝐒𝐭𝐚𝐭𝐬_𝐑𝐞𝐟𝐥𝐞𝐜𝐭𝐢𝐨𝐧𝐬.𝐣𝐩𝐞𝐠: An image displayed in the HTML report, complementing the section on statistical and scientific reflections.
– 𝐬𝐭𝐚𝐭𝐪𝐮𝐨𝐭𝐞_𝐜𝐲𝐜𝐥𝐞_𝐬𝐭𝐚𝐭𝐞.𝐫𝐝𝐬: An RDS file that stores the state of the statistical quotes cycle. This file is dynamically updated to prevent repetitions, ensuring that the quotes presented in the report change with each execution.

5_𝐅𝐞𝐚𝐭𝐮𝐫𝐞 𝐏𝐫𝐢𝐨𝐫𝐢𝐳𝐞𝐫 – 𝐑 𝐌𝐚𝐫𝐤𝐝𝐨𝐰𝐧 𝐒𝐜𝐫𝐢𝐩𝐭
The "Feature Priorizer" is an R Markdown-based analytical pipeline (Script_Feature_Prioritizer.Rmd) developed to perform the full multivariate analysis workflow presented in the study. The script integrates:

A) Dimensionality reduction (Logistic PCA)
B) Unsupervised clustering (K-Means)
C) Feature prioritization using the Nihans Index and Pareto Analysis
D) Statistical and practical significance assessment (Chi-square test, Cohen's h)
E) Automated report generation in HTML format, including figures and tables

6_𝐅𝐢𝐥𝐞𝐬 𝐑𝐞𝐥𝐚𝐭𝐞𝐝 𝐭𝐨 𝐭𝐡𝐞 𝐅𝐞𝐚𝐭𝐮𝐫𝐞 𝐏𝐫𝐢𝐨𝐫𝐢𝐳𝐞𝐫
– 𝐒𝐜𝐫𝐢𝐩𝐭_𝐅𝐞𝐚𝐭𝐮𝐫𝐞_𝐏𝐫𝐢𝐨𝐫𝐢𝐭𝐢𝐳𝐞𝐫.𝐑𝐦𝐝: The R Markdown script that executes the entire analytical pipeline
– 𝐒𝐜𝐫𝐢𝐩𝐭_𝐅𝐞𝐚𝐭𝐮𝐫𝐞_𝐏𝐫𝐢𝐨𝐫𝐢𝐭𝐢𝐳𝐞𝐫.𝐡𝐭𝐦𝐥: The automatically generated HTML report containing all results, figures, and statistical summaries
– 𝐈𝐧𝐬𝐭𝐚𝐥𝐥_𝐩𝐚𝐜𝐤𝐚𝐠𝐞𝐬.𝐑𝐦𝐝: A helper script that installs all necessary R packages for running the Feature Priorizer

本仓库包含与研究《巴西临床实验室关键结果管理综述：基于多变量分析、优先级排序与“新统计”方法的实践画像》相关的补充材料。仓库内提供数据集、图表、导出结果与分析脚本，以确保研究成果的完全可复现性与透明性。

文件夹结构

1_数据集/
该文件夹包含本研究使用的数据集，可直接用于特征优先级器（Feature Priorizer）的R Markdown脚本。

2_图表/
所有由特征优先级器生成的图表均存储于此，分辨率为600 DPI，可满足出版与分析的高质量绘图需求。

3_导出结果/
该文件夹包含导出的结果，包括统计输出、表格与分析所得的处理后数据集。

4_补充文件/
该文件夹包含用于生成特征优先级器HTML报告的辅助文件，以提升可视化呈现效果并集成动态统计引语。
– styles.css：定义HTML报告的格式，确保视觉呈现的一致性。
– logo.html、logo.png、logo.txt：与项目视觉标识相关的文件。
– Science_Stats_Reflections.jpeg：HTML报告中展示的图片，用于补充统计与科学反思部分的内容。
– statquote_cycle_state.rds：存储统计引语循环状态的RDS文件，该文件会动态更新以避免重复，确保报告每次运行时展示的引语均不相同。

5_特征优先级器——R Markdown脚本
“特征优先级器”是一款基于R Markdown的分析流程（Script_Feature_Prioritizer.Rmd），用于执行本研究中完整的多变量分析工作流。该脚本集成了以下功能：
A) 降维（逻辑主成分分析，Logistic PCA）
B) 无监督聚类（K均值聚类，K-Means）
C) 基于尼汉斯指数（Nihans Index）与帕累托分析的特征优先级排序
D) 统计与实际显著性评估（卡方检验，Chi-square test、Cohen's h）
E) 自动生成包含图表与表格的HTML格式报告

6_与特征优先级器相关的文件
– Script_Feature_Prioritizer.Rmd：执行完整分析流程的R Markdown脚本
– Script_Feature_Prioritizer.html：自动生成的HTML报告，包含所有结果、图表与统计汇总
– Install_packages.Rmd：用于安装运行特征优先级器所需全部R包的辅助脚本